Subject: Minutes of July 14 User's Committee Meeting

From: strauss@astro.Princeton.EDU

Submitted: Tue, 15 Jul 1997 09:10:45 -0400 (EDT)

Message number: 154 (previous: 153, next: 155 up: Index)

   Minutes of APO 3.5m User's Committee Phone Conference
		  Monday, July 14, 1997

  Attending: Ed Kibblewhite, Julie Lutz, Rene Walterbos, Bruce
Gillespie, Ed Turner (Chair), Michael Strauss (taking minutes), Chris
Stubbs, Alan Uomoto not available.

  The scheduling of shutdowns
  Slip of August-September shutdown.
  Twilight time: who gets it? 
  Community meeting in September 
  Getting User Feedback at the end of a run
  Miscellaneous News
			Scheduling Engineering Shutdowns

Ed Turner sent around a plan for scheduling major engineering
shutdowns (as opposed to one-night engineering runs) in apo35-general
#152 (July 2).  It has been very difficult to accurately schedule
major shutdowns months in advance, and Ed's plan suggests simply
scheduling the telescope fully, and scheduling shutdowns as little as
two weeks in advance; engineering shutdown should be seen to be like
bad weather.  

Stubbs: UW view of scheduling shutdown: 
Generally, people are supportive of telescope getting fixed.  However,
there was no enthusiasm for getting stomped on.   People would *much*
rather have a schedule that we could all believe.  Unfortunately, this
would probably require a full-time project manager.  Another
suggestion: shut the telescope down completely for a quarter, and fix
*everything*.  In practice, this would take many more people than we
have available. 
  The problem is partly psychological: If you have time scheduled, and
it is pre-empted for engineering maintenance work (as opposed to
fixing some emergency that keeps you from working), you can always
argue that they should wait a few days until your program is done, and
boot the *next* observer off the telescope. 

   Many ideas were traded back and forth, with no consensus as to the
best approach. 
  A possibility: telescope scheduled by institution, not by
individuals.  Do the individual scheduling, say, 2 weeks in advance.
This is quite a radical change for solving a very specific problem.
Especially for places with small amount of time (e.g., Washington
State).  What about synoptic projects?  In general, the user's
committee was unenthusiastic about block scheduling.
  A possibility: schedule an engineering shutdown period at the end of
each quarter, then if you decide you need a shutdown earlier than
that, you switch people around. 
  A risk: if you schedule shutdowns only when everyone is ready,
you'll *never* be completely ready!  You'll never get going!
  Why not simply label those programs close to a shutdown that their
time might be rescheduled or cancelled? 
  Part of the problem is that you can't reschedule a person who loses
time to a shutdown for a full quarter, at which time their object may
no longer be reachable. 
  How about scheduling once a month? This is the way things were done
earlier, but it made things just too fragmented. 
  How about scheduling quarterly, but iterating the schedule further
each month, with feedback from each institution? 

  Strauss: none of these approaches will work for enormous shutdowns
like the 5.5 week shutdown coming next month!

Ed will try to put this all together and post a compromise. 

	Scheduling of Twilight Time:
  There is a perception in the community that it is an imposition on
the observatory to ask to use twilight time; after all, the schedule
only includes the time between evening and morning astronomical

  The current policy is as follows: the twilight time is offered to first
(or last) program of the night, unless engineering needs to be done
during that time (a problem roughly 10% of the time).
Thus this time is available.   However, the observing specialists are
not in a position to do your twilight calibrations (e.g., flats) for
you; you have to be there to do the work. 

  Ed Turner's quasar monitoring program is often the first or last
program of the night; it therefore formally owns the corresponding
twilight time, although it doesn't always use it, and the other
programs of the night can use it, especially if the instrument in
question is DIS. 

  Gillespie: let's not schedule twilight formally; this brings up too
many problems.

	  APO User's Meeting September 19-20 at APO
  Rene Walterbos is putting it together, and will post a detailed
  There is no agenda yet, but it will cover the 3-year plan progress 
(cf.,, and
science that people are doing with the telescope. 

          The effect of the August-September Shutdown slip
The schedule for the August-September shutdown slipped several weeks
(thus prompting the discussion on scheduling these shutdowns, above)
(apo35-general #148, 149).  This has caused a number of users to be
rescheduled.  Ed Turner has gotten very little feedback on this; is
everyone able to live with this?  Let him know ASAP if not. 

	User Feedback at the end of the Run
Strauss suggested that there be a way for the observer to enter a
report for a night or a run, saying how things went, both good and
bad, from the observer's point of view.  Is this something that should
be included in the night logs?  
  Filing such a report would allow:
    1. Frustrations to be vented;
    2. The successes of the observatory to be disseminated;
    3. Recurrent subtle problems to be diagnosed;
    4. The observer's perspective to be monitored. 

There already exists a problem reports web page at APO, which most
observers are probably unaware of: 
This is appropriate for problems needing specific engineering
attention.  We thus probably need another way to post more general
comments, some of which may be positive. 
Apparently Paula Szkody has thought about this previously; Strauss will contact her. 

	Miscellaneous News from Gillespie:
	  New air slits are being made for DIS
	  Alan Watson is procuring new (narrow-band?) filters for GRIM. 
	  The Fermilab Drift Scan Camera has officially been retired. 
	  Jeff Sundval (sp) from Chicago, came by starting to reconnoiter for
	  the installation of the echelle. 

  Last month's meeting minutes approved. 

  Next meeting is Monday, September 8, 1997. 12:30 EDT; we will skip a
APO APO APO APO APO  Apache Point Observatory 3.5m  APO APO APO
APO  This is message 154 in the apo35-general archive. You can find
APO  the archive on
APO  To join/leave the list, send mail to
APO  To post a message, mail it to